Post on 14-May-2023
Myotonic dystrophy type 1 associated with nuclear foci of mutant RNA, sequestration of muscleblind
proteins, and deregulated alternative splicing in neurons
Hong Jiang1, Ami Mankodi1, Maurice S. Swanson2, Richard T. Moxley1,
and Charles A. Thornton1
1Department of Neurology, University of Rochester School of Medicine and Dentistry, Box 673, 601
Elmwood Avenue, Rochester, NY 14642.
2Department of Molecular Genetics and Microbiology, Powell Gene Therapy Center, University of
Florida College of Medicine, Gainesville, FL 32610.
Corresponding Author:
Charles A. Thornton, MD
Department of Neurology Box 673 Room 5-5207
601 Elmwood Avenue
University of Rochester Medical Center
Rochester, NY 14642
Tel 585 275 2542
Fax 585 273 1255
Email Charles_Thornton@urmc.rochester.edu
Copyright © 2004 Oxford University Press
HMG Advance Access published October 20, 2004 by guest on M
ay 30, 2013http://hm
g.oxfordjournals.org/D
ownloaded from
ABSTRACT
Myotonic dystrophy type 1 (DM1) is caused by expansion of a CTG repeat in the DMPK gene. In
skeletal muscle, DM1 may involve a novel, RNA-dominant disease mechanism in which transcripts from
the mutant DMPK allele accumulate in the nucleus and compromise the regulation of alternative splicing.
Here we show evidence for a similar disease mechanism in brain. Examination of post-mortem DM1
tissue by fluorescence in situ hybridization indicates that the mutant DMPK mRNA, with its expanded
CUG repeat in the 3’ untranslated region, is widely expressed in cortical and subcortical neurons. The
mutant transcripts accumulate in discrete foci within neuronal nuclei. Proteins in the muscleblind family
are recruited into the RNA foci and depleted elsewhere in the nucleoplasm. In parallel, a subset of
neuronal pre-mRNAs show abnormal regulation of alternative splicing. These observations suggest that
CNS impairment in DM1 may result from a deleterious gain-of-function by mutant DMPK mRNA.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Introduction
Myotonic dystrophy type 1 (dystrophia myotonica, DM1) is the most common inherited disease of
skeletal muscle in adults. Muscle wasting and delayed relaxation after muscle contraction (myotonia) are
the prominent core features of this illness. However, the effects of DM1 on the nervous system, heart,
and ocular lens, while less obvious, also pose a major threat to function and survival. The CNS symptoms
of DM1 may include cognitive impairment, hypersomnolence, heightened sensitivity to anesthetic agents,
central hypoventilation, neuroendocrine dysfunction, and effects on personality and behavior [reviewed
by Harper (1) and Ashizawa (2)]. Some of these effects, such as, mental retardation in individuals with
congenital DM1, occur during development (3). Other symptoms, such as, hypersomnolence, appear
during adult life. The mechanism and neuropathologic correlates for CNS involvement in DM1 are
unknown.
The genetic basis for DM1 is a dominantly inherited unstable expansion of CTG triplet repeats in
the 3’ untranslated region of the myotonic dystrophy protein kinase (DMPK) gene (4). The severity of
DM1 roughly correlates with the length of the CTG repeat tract in peripheral blood (5). For example,
individuals with minimal expansions of 50-100 repeats generally have mild, late-onset symptoms,
whereas those with 1000 or more repeats usually have severe disease in infancy. In many tissues,
including skeletal muscle and brain, somatic instability leads to expansions that typically are 1,000 –
4,000 repeats in length (6,7).
Recent evidence suggests that the pathogenesis of skeletal myopathy in DM1 involves a toxic
gain-of-function by mutant RNA. The CUG repeat in the mutant transcript is highly expanded but the
portion of the mRNA encoding DMPK remains intact. CUG-expanded transcripts are retained in muscle
nuclei in focal inclusions (8,9). Accumulation of mutant RNA in the nucleus compromises the regulation
of alternative splicing for a subset of muscle transcripts (10-13), and may also disrupt regulation of
transcription (14). Several models for this RNA-mediated disease mechanism have been proposed. First,
the CUG expansion RNA may alter the activity or cellular distribution of CUG binding protein 1
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
(CUGBP1), a regulator of RNA splicing and translation (10,15). Second, the CUG expansion RNA may
form hairpin loop structures that interact with double-stranded RNA binding proteins (dsRBPs) (16).
Third, the CUG expansion RNA may interact with transcription factors, such as, retinoic acid receptor γ
and Sp1, depleting these factors from active chromatin (14).
A particularly attractive model for the disease mechanism in DM1 emerges from recent studies of
proteins in the muscleblind (MBNL) family (17). These proteins bind to expanded CUG repeats in vitro
and they are strongly recruited into RNA foci in muscle nuclei, raising the possibility the sequestration of
MBNL proteins leads to muscle disease in DM1 (17-20). In support of this model, disruption of murine
Mbnl1 produces a phenotype of myopathy (myotonia, defective regulation of alternative splicing) and lens
cataracts that closely resembles human DM1 (21).
However, it is unclear if RNA gain-of-function can explain all manifestations of DM1 in skeletal
muscle, or whether this mechanism can account for the non-muscle symptoms. Other effects of the DM1
mutation, such as, decreased levels of DMPK protein (due to nuclear retention of RNA from the mutant
allele) (8,9), or, silencing of the neighboring gene SIX5 (due to effects of the repeat expansion on
chromatin structure) (22-24) may also contribute to pathogenesis. Indeed, studies of knockout mice
suggest that reduced expression of genes at the DM1 locus contributes to the cardiac (in the Dmpk
knockouts) (25) or ocular (in the Six5 knockouts) (26,27) signs of DM1.
Presently it is unclear whether any steps in the pathogenic sequence of poly(CUG) expression,
formation of RNA inclusions, sequestration of RNA binding proteins, and disruption of alternative
splicing can take place in the CNS. There is controversy about which cells in the mature brain, if any,
express DMPK (28). To address these questions we have examined the expression and distribution of
expanded poly(CUG) RNA in relation to putative binding proteins in post-mortem brain tissue, and
screened for abnormalities of alternative splicing.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Results
Fluorescence in situ hybridization (FISH) of brain sections with CAG repeat probes revealed
nuclear RNA foci in every individual with DM1 (n=10, Fig. 1A) but not in controls with (n=7) or without
(n=6) neurologic disease. RNA foci were not observed with CUG (sense) or GUC repeat probes. The
hybridization of CAG probes to nuclear foci in DM1 did not require denaturation of genomic DNA.
These results indicate that CAG repeat probes recognize CUG expansion RNA rather than a cross-reactive
RNA or DNA.
Nuclear RNA foci ranged in diameter from 0.2 to 2 µm. Resolution of these small structures
required direct fluorescence detection methods. However, the autofluorescent material in brain
(lipofuscin) was a complicating factor. We found that RNA foci were clearly distinguished from
lipofuscin when the epifluorescence from three color channels was merged in a single image. As shown
in Fig. 1A, the nuclear foci appeared in a single channel determined by the probe label (Texas red).
Lipofuscin, which excites and emits at a broad spectrum of wavelengths, generated signal in all channels
and appeared as a different color (typically yellow-brown) in the merged image. These observations
formed the basis for distinguishing RNA foci in subsequent experiments. RNA foci were red, sharply-
demarcated structures in the nucleus. Lipofuscin was yellow-brown perinuclear material with indistinct
margins.
To determine which cells express mutant DMPK and form RNA inclusions we surveyed different
brain regions using FISH in combination with antibodies that mark specific cell types. In cerebral cortex
the nuclear foci were distributed throughout all cortical layers and were confined to neurons, as
determined by immunofluorescence (IF) for neuronal markers NeuN (Fig. 1B) or MAP2 (Fig. 1C).
Counts of 100 NeuN-positive cells from temporal and frontal cortex of 4 patients with classical DM1
(selected for best relative preservation of cortical architecture) showed RNA foci in >85% of cortical
neurons in each case. More than one focus was visible in ~30% of cortical neurons, and occasional
neurons had up to 15 small foci. In contrast, the individual having a small CTG repeat expansion (77
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
repeats) and mild phenotype (cataracts, mild weakness, and cognitive impairment after age 60 years) had
foci in only 39% of NeuN-positive neurons in temporal cortex.
RNA foci were widely distributed in other neuronal populations, including the hippocampus (all
sectors), dentate gyrus, thalamus, and also the substantia nigra and brain stem tegmentum (each of 4
patients examined) (Fig. S1). The main exception was in cerebellar cortex, where small foci were
detected in some Purkinje cells but not in neurons of the molecular or granular cell layers (n=6 patients
examined) (Fig. 1D). RNA foci were also present in the subcortical white matter and corpus callosum in
occasional cells expressing 2’3’-cyclic nucleotide 3’-phosphodiesterase (CNPase), a marker for
oligodendrocytes (Fig. 1E). However, these foci were smaller and less intense than those in cortical
neurons. In sections processed on the same slide and imaged under the same exposure settings,
quantitation of FISH signals indicated that the amount of CUG expansion RNA in frontal cortical neurons
was 2.9-fold greater (area × intensity) than in Purkinje cells (p<10-10) and 18-fold greater than in
oligodendrocytes (p<10-10) within the same individual (n=3 patients, 60 nuclei per patient).
Paired samples of frontal cortex and biceps muscle were available for three patients. When
sections of skeletal muscle and cerebral cortex from same patient were processed on the same slide and
imaged under the same exposure settings, the RNA inclusions were larger and more intense (3.1-fold
greater, area × intensity) in frontal cortical neurons than in skeletal muscle from the same individual
(p<10-10, Fig. S2).
To determine if mutant RNA resides in a previously identified nuclear domain, we tested for
colocalization of mutant RNA with proteins that mark different nuclear compartments. These and
subsequent experiments localizing protein relative to expanded poly(CUG) RNA were performed on a
subset of 4 DM1 and 3 non-disease control samples showing the best preservation of cortical architecture.
In contrast to nuclear inclusions of polyglutamine proteins (29), RNA foci did not colocalize with PML
bodies (Fig. 1F). We also did not find colocalization of mutant RNA with the nucleolus (visualized by
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
DNA staining or antibodies to C23 nucleolin), perinucleolar compartment (antibodies to polypyrimidine
tract binding protein), or “speckles” (antibodies to hnRNP C) (data not shown). We cannot eliminate the
possibility of colocalization with Cajal bodies because p80 coilin antibodies did not consistently identify
Cajal bodies in cortical neurons stained by our methods.
The proteasome and exosome are multisubunit complexes responsible for protein and RNA
degradation, respectively. To determine if these complexes are recruited to nuclear RNA foci, we
combined FISH with immunofluorescence using antibodies to components of the proteasome or exosome.
Three components of the proteasome (20Sα, 11Sγ and 11Sα subunits) were recruited to RNA foci in
cortical neurons (Fig. 2A; Fig. S3A). We did not, however, find evidence for ubiquitination or
sumoylation of the foci (not shown). In contrast, antibodies to the PM/Scl75 or PM/Scl100 components
of the exosome did not colocalize with RNA foci (Fig. 2B)
Monoclonal antibody 3B1 showed strong expression of CUGBP1 in cortical neurons (Fig. 2C).
The distribution of this protein in neuronal nucleus and cytoplasm appears similar in DM1 patients and
controls, and FISH/IF analysis shows that CUGBP1 is not recruited into RNA foci. Polyclonal antibodies
to other members of the CUGBP1 family, ETR3 and CELF4, also fail to colocalize with foci (not shown).
None of six different dsRNA binding proteins in neuronal nuclei (staufen, NF90, ADAR1, PACT, PKR,
RNA helicase A) colocalize with RNA foci (representative images for NF90 are shown in Fig. 2D and
ADAR1 in Fig. S3B). The RNA binding proteins hnRNP A1, hnRNP I, hnRNP M, KSRP and HuR did
not colocalize with RNA foci (representative image for hnRNP M is shown in Fig S3D). In contrast,
hnRNPs H and F colocalized with foci in cortical neurons to a limited extent (Fig. 2F, Fig. S3C), and
these results were verified using two different polyclonal antibodies for each protein. The intensity of
immunofluorescence for these proteins was greatest at the site of RNA foci; however, there did not appear
to be significant depletion of hnRNP H or hnRNP F elsewhere in the neuronal nucleoplasm.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Mutant DMPK mRNA is reported to interact with transcription factors retinoic acid receptor
gamma (RARγ) and Sp1 (14). In cortical neurons, these transcription factors were readily detected by
immunofluorescence but they did not colocalize with RNA foci (Figs. 2I and S3E) and their distribution
was similar in DM1 patients and controls (Fig. 2I and 2J).
Polyclonal antisera recognizing all members of the muscleblind family (MBNL1, MBNL2, and
MBNL3) showed strong colocalization with RNA foci (not shown). We used monoclonal antibodies
raised against epitopes specific for MBNL1 or MBNL2 to determine which muscleblind proteins interact
with CUG expansion RNA in neurons. MBNL3 was not examined because its expression in adults is
mainly restricted to placenta (20,30). In normal controls, monoclonal antibody 3A4 showed expression of
MBNL1 in nuclei and cytoplasm of cortical neurons (Fig. 2H). In DM1, MBNL1 was strongly recruited
into RNA foci whereas staining elsewhere in the nucleus was markedly reduced (Fig. 2G). Quantitative
analysis was performed on 3 DM1 patients and non-neurologic disease controls having the shortest post-
mortem intervals and best preservation of cortical architecture (Fig. 3). The mean immunofluorescence
intensity for MBNL1 in the nucleoplasm (excluding RNA foci and nucleoli) was 2.3-fold lower in DM1
neurons than in non-disease controls (26 ± 9 area × intensity units in DM1 patients vs 61 ± 17 in controls,
20 neuronal nuclei per subject, p<0.00001). Monoclonal antibody 2D9 showed that MBNL2 was also
recruited into RNA foci (Fig. 2E). However, immunofluorescence signals in neurons with MoAb 2D9
were lower in relation to background staining in the neuropil, precluding a reliable quantification of its
distribution.
To determine if DM1 is associated with altered regulation of alternative splicing in brain, we
examined 45 exons (in 31 genes) known to undergo alternative splicing in brain (supplemental Table 2).
For each exon, the ratio of inclusion versus exclusion isoforms was determined by reverse transcriptase-
PCR (RT-PCR) using primers flanking the regulated exon. An initial screen was performed using total
RNA extracted from superior temporal cortex from two controls without neurological disease and four
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
DM1 patients. Among 45 exons screened, 4 appeared to show a change in the ratio of exon
inclusion/exclusion splice products in DM1. These differences were then confirmed and quantified in
triplicate assays using temporal cortex RNA from 7 patients with DM1 and 5 controls (Fig. 4). DM1 was
associated with decreased inclusion of amyloid precursor protein exon 7 (10 ± 1% in DM1, 30 ± 11% in
controls, p<0.001), increased inclusion of NMDA NR1 receptor exon 5 (33 ± 11% in DM1, 11 ± 5% in
controls, p<0.01), decreased inclusion of tau exon 2 (5 ± 1% in DM1, 36 ± 10% in controls, p<10-5), and
decreased inclusion of tau exon 10 (21 ± 1% in DM1, 41 ± 5% in controls, p<10-6).
Discussion
The DM1 mutation has complex effects on genome function. Expansion of the CTG repeat in
DMPK leads to reduced expression of DMPK protein (due to nuclear retention of the mutant mRNA),
trans-interference with alternative splicing (10), and partial silencing of genes at the DM1 locus, such as,
the transcription factor SIX5 (due to effects on chromatin structure) (22-24). Following a reductionist
approach, each of these effects has been separately modeled in mice to assess their respective roles in
pathogenesis (26,27,31-34). It is unclear, however, if any models have reproduced the CNS symptoms of
DM1. Neurobehavioral studies have not been reported for Dmpk, Six5, or Mbnl1 knockout mice. Dmpk
knockout mice show abnormalities of hippocampal physiology (35). However, this finding has no known
correlate in DM1, in which there is only partial DMPK deficiency. Transgenic mice expressing CUG
expansion RNA in the brain show an abnormal distribution of tau protein isoforms (34). However, the
pattern of transgene expression in the CNS has not been determined, and it is unknown whether these
mice develop RNA foci in neurons, abnormal splicing of tau pre-mRNA, or CNS impairment.
Neuronal intranuclear inclusions are characteristic of several neurological disorders. In the
polyglutamine disorders, the core component of the inclusion is mutant protein or a cleavage product
containing the polyglutamine tract (36,37). In neuronal intranuclear inclusion disease, the main
component of the inclusion has not been identified (38). In Fragile X tremor ataxia syndrome (FXTAS),
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
FMR1 mRNA having an expanded CGG repeat leads to formation of nuclear inclusions (39). Our results
indicate that DM1 should be added to the list of disorders characterized by neuronal intranuclear
inclusions. In general terms, it is uncertain whether nuclear inclusions directly contribute to neuronal
dysfunction in any of these disorders (40,41). In DM1 muscle tissue, however, evidence suggests that
RNA inclusions are directly involved in disease pathogenesis, through a mechanism that involves
sequestration of muscleblind proteins and mis-regulation of alternative splicing (21,33). Our observations
showing strong expression of expanded poly(CUG) RNA in DM1 neurons, formation of RNA inclusions,
redistribution of muscleblind proteins, and altered regulation of alternative splicing raise the possibility
that CNS symptoms of DM1 also are triggered by RNA inclusions.
The phenotypes that distinguish different CAG•CTG expansion disorders reflect distinct patterns
of cell vulnerability. For example, the polyglutamine proteins in Huntington’s disease or the
spinocerebellar ataxias are widely expressed yet the phenotypes are almost exclusively due to neuronal
dysfunction. DM1 presents a different pattern. Despite evidence that mutant DMPK RNA accumulates to
higher levels in cortical neurons (Fig S2), the cell degeneration is more severe in muscle. Our initial
studies also suggest that splicing abnormalities are less frequent and less severe in cerebral cortex than in
skeletal muscle (Fig 4 and unpublished observations, C Thornton and X Lin), raising the possibility that
muscleblind proteins are more effectively sequestered in muscle nuclei, or that compensation for
muscleblind deficiency is more effective in neurons, perhaps due to expression of additional RNA binding
proteins. The exact determinants of cell vulnerability in DM1 are unknown but the stoichiometry of CUG
expansion RNA in relation to muscleblind proteins is likely to play an important role. For this reason,
differences in the size of foci in different cells do not necessarily predict severity of an RNA-mediated
disease process.
In spinocerebellar ataxia type 8 (SCA8) an expanded CTG repeat in a gene expressing a non-
coding RNA can lead to progressive cerebellar dysfunction (42). As in DM1, a CUG expansion RNA is
presumably expressed in SCA8, but the CNS symptoms of these disorders show little overlap. The lack
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
of RNA foci in DM1 cerebellar cortex, except for small inclusions in Purkinje cells, might account for
this difference, but other factors, such as, length of the CUG expansion or flanking sequence in the mutant
RNA, may also influence the pattern of neuronal vulnerability.
Initial studies of DMPK immunolocalization showed expression in cortical neurons or ependymal
cells (43,44), but questions about antibody specificity were raised (45). A more recent study used a panel
of 16 different monoclonal antibodies to rigorously establish antibody specificity (46). DMPK protein
was found exclusively in skeletal, cardiac, and smooth muscle. Lacking evidence for DMPK expression
in brain, the authors postulated that CNS symptoms of DM1 result from haploinsufficiency for genes at
the DM1 locus. The present studies, however, establish that the mutant DMPK mRNA is widely
expressed in cortical and subcortical neurons. Expression in specific neuronal populations is consistent
with Northern blots showing that levels of DMPK mRNA are lower in brain than in skeletal or cardiac
muscle (4) and a recent paper showing wide expression of DMPK in cortical neurons, assessed by in situ
hybridization in mice (47). The failure to detect DMPK immunologically likely reflects its relatively low
concentration in brain homogenates.
The most conspicuous effect on alternative splicing that we observed in DM1 brain was a 3-fold
increase in the fraction of NMDA receptor 1 (NMDAR1) mRNA that includes exon 5. Inclusion of this
exon influences the pharmacologic behavior, gating, and cellular distribution (somatic rather than somato-
dendritic expression) of NMDAR1 (48-50). NMDAR1 function is required for normal long term
potentiation in the hippocampus and learning (51,52). It seems possible, therefore, that altered splicing of
exon 5 may contribute to the memory impairment observed in DM1 (53,54).
Microtubule-associated protein tau (MAPT) pre-mRNA is alternatively spliced at exons 2, 3, and
10 (55). Tau transcripts in fetal brain do not include exon 10, whereas ~50% of transcripts in adult brain
include this exon which encodes an additional microtubule binding domain (55-57). Alternative splicing
of exons 2 and 3 also is developmentally regulated (neither exon is included in the fetus, adults mainly
include exon 2). The relative proportion of tau splice products is tightly regulated, as shown by kindreds
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
with frontotemporal dementia and parkinsonism (FTDP-17) due to mutations in MAPT. Silent mutations
in MAPT exon 10, or, in the flanking intron, lead to FTDP-17 by disrupting cis elements that regulate
splicing of tau pre-mRNA (58-61). Usually these mutations lead to increased inclusion of exon 10 (62).
However, some MAPT mutations that segregate with FTDP-17 have the opposite effect of reducing exon
10 inclusion (61,63). Notably, DM1 is associated with reduced exon 10 inclusion (Fig 4), and FTDP-17
and DM1 are both associated with neurofibrillary tangles and neuronal aggregates of hyperphosphorylated
tau (64-67). We also found that inclusion of tau exon 2 is reduced in DM1, confirming previous
observations (68). Our results predict that fetal isoforms of tau (excluding exons 2, 3, and 10) are
inappropriately expressed in adult DM1 brain, findings that correlate well with previous studies of tau
protein in DM1 brain(68). Expression of human fetal tau in transgenic mice leads to formation of
neurofibrillary tangles and axonopathy (69,70). It is unclear, however, whether the extent of the tau mis-
splicing in DM1 is sufficient to cause neuronal dysfunction.
We have also found that DM1 is associated with increased expression of fetal splice isoforms for
APP (exon 7 exclusion products). Taken together, these observations suggest that accumulation of
mutant DMPK mRNA in the neuronal nucleus compromises a specific developmental program of
alternative splicing.
Among the RNA binding proteins that we examined, only MBNL1 and MBNL2 are strongly
recruited to nuclear foci of mutant RNA. In Drosophila, muscleblind is required for terminal
differentiation of muscle and photoreceptor cells (71). Human orthologues of muscleblind were initially
isolated as the major poly(CUG) binding proteins in HeLa nuclear extracts (17). The three mammalian
muscleblind genes, MBNL1, MBNL2, and MBNL3, are closely related (20,30). When expressed as GFP
fusion proteins, each member of this family can localize to RNA foci in DM1 cells (20). Our results are
the first to show depletion of MBNL1 in the nucleoplasm of DM1 cells, supporting a model where CUG
expansion RNA accumulates to levels sufficient to sequester and compromise the nuclear functions of
MBNL1.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Loss of Mbnl1 is sufficient to induce DM1-like defects of alternative splicing in murine striated
muscle (21). A recent paper has shown that human MBNL1 is a direct regulator of alternative splicing
(72). RNA-binding proteins in the CELF family (CUG binding protein and ETR3-like factors) are also
implicated in DM1-related defects of RNA processing (10,11,73). The initial observations linking CELF
proteins with DM1 pathogenesis indicated that CUG binding protein 1 (CUGBP1) and ETR3/CUGBP2
bind to short oligo(CUG) RNAs in vitro (15,74). Subsequent studies, however, failed to show interaction
of CELF proteins with expanded poly(CUG) in vitro or in DM1 cells (19,75,76). Proteins in this family
nevertheless regulate splicing of several exons that show defective regulation in DM1 muscle (10,11,13).
Notably, CELF proteins may also regulate alternative splicing for tau exon 10 and NMDAR1 exon 5
(77,78). However, it is unlikely that sequestration of CELF proteins on CUG expansion RNA is
responsible for splicing defects in neurons because monoclonal antibody 3B1, which recognizes an
epitope shared by several CELF proteins (79), does not colocalize with RNA inclusions. Moreover, the
splicing defects that we observed would predict opposite effects on CELF protein activity [decreased
activity of CELF3 or CELF4 in the case of tau exon 10 (78), or increased activity of ETR3/CUGBP2 in
the case of NMDAR1 exon 5 (77)]. Our results do not suggest a unifying mechanism whereby CUG
expansion RNA directly alters the activity of CELF proteins. It remains possible, however, that nuclear
accumulation of mutant RNA or sequestration of muscleblind proteins may indirectly affect the activity of
specific CELF proteins in the neuronal nucleus.
The mutant DMPK mRNA in the RNA inclusion is, at least in part, the full-length mRNA (8,9).
Nuclear mRNAs exist as ribonucleoprotein (RNP) complexes rather than naked polynucleotides (80).
Thus, RNA inclusions would be expected to contain other nuclear RNA binding proteins in addition to
MBNL1 and MBNL2. We found that hnRNPs H and F (but not hnRNP C, hnRNP I, hnRNP M, KSRP,
or HuR) colocalized with nuclear foci of poly-CUG RNA to a limited extent. However, the overall
nuclear distribution of hnRNP H or F in DM1 neurons did not show an obvious change. Also, the
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
splicing of neuron-specific exon N1 of c-src, which is promoted by hnRNPs H and F (81,82), was not
reduced in DM1 cerebral cortex. Indeed, inclusion of the N1 exon showed a slight (1.3-fold, p<0.02)
increase in DM1 with respect to controls, opposite to the predicted effect of hnRNP F or H depletion (not
shown). This fits with expectations that the number and density of binding sites on a single transcript,
hence the capacity for protein sequestration, is much greater for proteins that bind to expanded
poly(CUG) than for proteins that bind DMPK mRNA outside of the repeat tract.
A recent study suggested that mutant DMPK transcripts interact with transcription factors Sp1 and
RARγ, “leaching” these factors away from normal binding sites on chromatin (14). At present, however,
there is no direct biochemical evidence for interaction between poly(CUG) and DNA binding proteins.
We have not found that Sp1 or RARγ colocalize with RNA inclusions in DM1 neurons. A caveat is that
post-mortem autolysis, loss of architectural preservation in frozen tissue, or masking of epitopes may
have limited our ability to detect RNA-protein interactions.
Factors that control the accumulation and degradation of mutant DMPK mRNA in the nucleus are
not understood. In the polyglutamine disorders, the proteasome is recruited to nuclear foci of mutant
protein (83). We postulated that the exosome, a multisubunit complex responsible for mRNA degradation
(84), would be recruited to nuclear foci of mutant RNA in DM1. Opposite to this prediction, we found
that the proteasome is recruited to RNA foci whereas the exosome is not. This observation raises the
possibility that the proteasome is recruited by conformational changes in MBNL1, MBNL2, or other
poly(CUG) binding proteins. If this is the case, loss of muscleblind function in DM1 may result from the
combined effects of sequestration and accelerated degradation.
Materials And Methods
Tissue samples.
Autopsy materials were obtained from ten DM1 patients (mean age 56 years, range 44-78 years, 7
men and 3 women) and 13 controls (6 with no neurologic disease, 2 with Alzheimer disease, 4 with
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Huntington disease, and one with refractory epilepsy). The mean post-mortem interval for DM1 patients
was 6 hours (range 2 to 14 hours). At the time of autopsy coronal sections of brain were prepared and
placed on aluminum slabs cooled on dry ice. In addition, selected regions were dissected and flash frozen
in liquid nitrogen. All samples were stored at -70°C. Nine of the DM1 patients had signs of classical
DM1 before age 30 and died of complications related to the disease (respiratory failure in 7, sudden
cardiac death in 2). The other DM1 patient had minimal symptoms of DM1 and died at age 78 yrs of
unrelated disease. Genetic confirmation was performed as previously described by PCR or Southern blot
on DNA isolated from postmortem brain tissue (7). Southern blots of cortical DNA samples showed a
broad range of expanded alleles ranging in size from 5 to 12 kb (not shown). The individual with the
minimal DM phenotype had a CTG repeat expansion length of 77 repeats in DNA isolated from
peripheral blood, brain, and other tissues.
Fluorescence in situ hybridization (FISH)
FISH was performed as described (18) with slight modifications. Frozen sections (12 µm) were
fixed in 3% paraformaldehyde PBS for 30 min, permeabilized in 2% acetone PBS (pre-chilled at -20°C)
for 5 min, and then prehybridized in 30% formamide and 2 X SSC at room temperature for 10 min. Next,
sections were hybridized with probe (1 ng/µl) for 2 h at 37°C in buffer (30% formamide, 2 X SSC, 0.02
% BSA, 66 µg/ml yeast rRNA, 2 mM vanadyl complex) and then washed for 30 min in 30%
formamide/2XSSC at 42°C followed by 1X SSC for 30 min at room temperature. Probes were HPLC-
purified 2-O-methyl RNA 20-mers (IDT, Coralville, IA) composed of CAG-, CUG- or GUC- repeats, and
labeled with Texas Red at the 5’ end. Images were obtained on an Olympus AX70 epifluorescence
microscope at 1,000-fold magnification. To compare the relative fluorescence intensities for RNA foci,
sections were processed on the same slide, imaged under the same illumination and exposure settings, and
then analyzed using MCID v6.0 software (Imaging Research Inc., St. Catherines, Ontario).
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Immunofluorescence (IF) combined with FISH
Following the 1 X SSC post-hybridization wash of the FISH procedure, sections were incubated in
primary antibodies (listed in supplemental Table 1) overnight at 4°C, washed five times with PBS for 2
min, and then incubated in secondary antibody (Alexa 488-labeled goat anti-rabbit polyclonal or Alexa
488-labeled goat anti-mouse polyclonal, Molecular Probes) and 33 nM diamidino-2-phenylindole (DAPI)
for 30 min at room temperature. Sections were washed five times in PBS prior to mounting. To estimate
relative MBNL1 concentration in nucleoplasm in DM1 nuclei vs controls, sections of temporal cortex
were processed on the same slide and imaged under the same exposure settings. Merged images for Texas
red (to visualize RNA foci), Alexa 488 (for MBNL1) and DAP1 (for nuclear DNA) were obtained.
Regions of interest were manually defined as nuclear area excluding nucleolus, RNA foci, and
overlapping lipofuscin. MBNL1 fluorescence intensity (mean optical density in monochrome mode in
arbitrary units) in the region of interest was determined for 20 cortical neuronal nuclei per subject.
Because of the difficulty estimating background fluorescence from brain sections, the results are not
corrected for background. This approach provides a conservative estimate of the fold-reduction for
MBNL1 in DM1 nucleoplasm.
Splicing assays
Alternative splicing was assessed by RT-PCR. Total RNA was isolated from temporal cortex gray
matter of 7 DM1 patients and 5 non-neurologic disease controls using TriReagent (Molecular Research
Center, Cincinnati). cDNA was synthesized using SuperScript II reverse transcriptase (Invitrogen) with a
mixture of oligo(dT)12-18 and random hexamer primers. The cDNA was digested with RNase H and then
amplified using PCR primers flanking alternatively spliced exons (Table 2). PCR products were resolved
on agarose gels, stained with SybrGreenII (Molecular Probes), and analyzed on a fluorimager. An initial
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
screen was performed on a subset of samples (4 DM1 and 2 control). Four exons appeared to show
deregulated splicing in DM1. These differences were quantified in a second experiment including the full
panel of 7 DM1 and 5 control samples. The fraction of exon inclusion was determined on triplicate
reactions using ImageQuant software (Amersham, Piscataway).
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
ACKNOWLEDGEMENTS
The authors thank Don Henderson for excellent technical assistance. This work was supported by
NIH/NIAMS AR49077, AR48143 (C.T.), and AR46799 (M.S.S.), the Muscular Dystrophy Association,
and the Saunders Family Neuromuscular Research Fund. The authors thank the University of Rochester
Alzheimer Disease Center for tissue samples and Drs Thomas Cooper, Jeffrey Wilusz, Doug Black,
Ganes Sen, C.Lee, D. Cho, KL Chan, E. Wagner for gifts of antibodies. This work comes from the
University of Rochester Paul D. Wellstone Muscular Dystrophy Cooperative Research Center
(AR050762).
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
REFERENCES
1. Harper,P.S. (2001) Myotonic dystrophy. Saunders, London.
2. Ashizawa,T. (1998) Myotonic dystrophy as a brain disorder. Arch.Neurol., 55, 291-293.
3. Dyken,P.R., Harper,P.S. (1973) Congenital dystrophia myotonica. Neurology, 23, 465-473.
4. Brook,J.D., McCurrach,M.E., Harley,H.G., Buckler,A.J., Church,D., Aburatani,H., Hunter,K.,
Stanton,V.P., Thirion,J.P., Hudson,T. et al. (1992) Molecular basis of myotonic dystrophy:
expansion of a trinucleotide (CTG) repeat at the 3' end of a transcript encoding a protein kinase
family member. Cell, 68, 799-808.
5. Harley,H.G., Rundle,S.A., MacMillan,J.C., Myring,J., Brook,J.D., Crow,S., Reardon,W.,
Fenton,I., Shaw,D.J., and Harper,P.S. (1993) Size of the unstable CTG repeat sequence in relation
to phenotype and parental transmission in myotonic dystrophy. Am.J.Hum.Genet., 52, 1164-1174.
6. Ashizawa,T., Dubel,J.R., and Harati,Y. (1993) Somatic instability of CTG repeat in myotonic
dystrophy. Neurology, 43, 2674-2678.
7. Thornton,C.A., Johnson,K., and Moxley,R.T. (1994) Myotonic dystrophy patients have larger
CTG expansions in skeletal muscle than in leukocytes. Ann.Neurol., 35, 104-107.
8. Taneja,K.L., McCurrach,M., Schalling,M., Housman,D., and Singer,R.H. (1995) Foci of
trinucleotide repeat transcripts in nuclei of myotonic dystrophy cells and tissues. J.Cell Biol., 128,
995-1002.
9. Davis,B.M., McCurrach,M.E., Taneja,K.L., Singer,R.H., and Housman,D.E. (1997) Expansion of
a CUG trinucleotide repeat in the 3' untranslated region of myotonic dystrophy protein kinase
transcripts results in nuclear retention of transcripts. Proc.Natl.Acad.Sci.U.S.A, 94, 7388-7393.
10. Philips,A.V., Timchenko,L.T., and Cooper,T.A. (1998) Disruption of splicing regulated by a
CUG-binding protein in myotonic dystrophy. Science, 280, 737-741.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
11. Savkur,R.S., Philips,A.V., and Cooper,T.A. (2001) Aberrant regulation of insulin receptor
alternative splicing is associated with insulin resistance in myotonic dystrophy. Nat.Genet., 29, 40-
47.
12. Mankodi,A., Takahashi,M.P., Jiang,H., Beck,C.L., Bowers,W.J., Moxley,R.T., Cannon,S.C., and
Thornton,C.A. (2002) Expanded CUG repeats trigger aberrant splicing of ClC-1 chloride channel
pre-mRNA and hyperexcitability of skeletal muscle in myotonic dystrophy. Mol.Cell, 10, 35-44.
13. Charlet,B., Savkur,R.S., Singh,G., Philips,A.V., Grice,E.A., and Cooper,T.A. (2002) Loss of the
muscle-specific chloride channel in type 1 myotonic dystrophy due to misregulated alternative
splicing. Mol.Cell, 10, 45-53.
14. Ebralidze,A., Wang,Y., Petkova,V., Ebralidse,K., and Junghans,R.P. (2004) RNA leaching of
transcription factors disrupts transcription in myotonic dystrophy. Science, 303, 383-387.
15. Timchenko,L.T., Miller,J.W., Timchenko,N.A., DeVore,D.R., Datar,K.V., Lin,L., Roberts,R.,
Caskey,C.T., and Swanson,M.S. (1996) Identification of a (CUG)n triplet repeat RNA-binding
protein and its expression in myotonic dystrophy. Nucleic Acids Res., 24, 4407-4414.
16. Tian,B., White,R.J., Xia,T., Welle,S., Turner,D.H., Mathews,M.B., and Thornton,C.A. (2000)
Expanded CUG repeat RNAs form hairpins that activate the double- stranded RNA-dependent
protein kinase PKR. RNA., 6, 79-87.
17. Miller,J.W., Urbinati,C.R., Teng-Umnuay,P., Stenberg,M.G., Byrne,B.J., Thornton,C.A., and
Swanson,M.S. (2000) Recruitment of human muscleblind proteins to (CUG)(n) expansions
associated with myotonic dystrophy. EMBO J., 19, 4439-4448.
18. Mankodi,A., Urbinati,C.R., Yuan,Q.P., Moxley,R.T., Sansone,V., Krym,M., Henderson,D.,
Schalling,M., Swanson,M.S., and Thornton,C.A. (2001) Muscleblind localizes to nuclear foci of
aberrant RNA in myotonic dystrophy types 1 and 2. Hum.Mol.Genet., 10, 2165-2170.
19. Fardaei,M., Larkin,K., Brook,J.D., and Hamshere,M.G. (2001) In vivo co-localisation of MBNL
protein with DMPK expanded-repeat transcripts. Nucleic Acids Res., 29, 2766-2771.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
20. Fardaei,M., Rogers,M.T., Thorpe,H.M., Larkin,K., Hamshere,M.G., Harper,P.S., and Brook,J.D.
(2002) Three proteins, MBNL, MBLL and MBXL, co-localize in vivo with nuclear foci of
expanded-repeat transcripts in DM1 and DM2 cells. Hum.Mol.Genet., 11, 805-814.
21. Kanadia,R.N., Johnstone,K.A., Mankodi,A., Lungu,C., Thornton,C.A., Esson,D., Timmers,A.M.,
Hauswirth,W.W., and Swanson,M.S. (2003) A muscleblind knockout model for myotonic
dystrophy. Science, 302, 1978-1980.
22. Otten,A.D., Tapscott,S.J. (1995) Triplet repeat expansion in myotonic dystrophy alters the
adjacent chromatin structure. Proc.Natl.Acad.Sci.U.S.A, 92, 5465-5469.
23. Klesert,T.R., Otten,A.D., Bird,T.D., and Tapscott,S.J. (1997) Trinucleotide repeat expansion at the
myotonic dystrophy locus reduces expression of DMAHP. Nat.Genet., 16, 402-406.
24. Thornton,C.A., Wymer,J.P., Simmons,Z., McClain,C., and Moxley,R.T. (1997) Expansion of the
myotonic dystrophy CTG repeat reduces expression of the flanking DMAHP gene. Nat.Genet., 16,
407-409.
25. Berul,C.I., Maguire,C.T., Aronovitz,M.J., Greenwood,J., Miller,C., Gehrmann,J., Housman,D.,
Mendelsohn,M.E., and Reddy,S. (1999) DMPK dosage alterations result in atrioventricular
conduction abnormalities in a mouse myotonic dystrophy model. J.Clin.Invest, 103, R1-R7.
26. Klesert,T.R., Cho,D.H., Clark,J.I., Maylie,J., Adelman,J., Snider,L., Yuen,E.C., Soriano,P., and
Tapscott,S.J. (2000) Mice deficient in Six5 develop cataracts: implications for myotonic
dystrophy. Nat.Genet., 25, 105-109.
27. Sarkar,P.S., Appukuttan,B., Han,J., Ito,Y., Ai,C., Tsai,W., Chai,Y., Stout,J.T., and Reddy,S.
(2000) Heterozygous loss of Six5 in mice is sufficient to cause ocular cataracts. Nat.Genet., 25,
110-114.
28. Lam,L.T., Pham,Y.C., Nguyen,T.M., and Morris,G.E. (2000) Characterization of a monoclonal
antibody panel shows that the myotonic dystrophy protein kinase, DMPK, is expressed almost
exclusively in muscle and heart. Hum.Mol.Genet., 9, 2167-2173.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
29. Skinner,P.J., Koshy,B.T., Cummings,C.J., Klement,I.A., Helin,K., Servadio,A., Zoghbi,H.Y., and
Orr,H.T. (1997) Ataxin-1 with an expanded glutamine tract alters nuclear matrix-associated
structures. Nature, 389, 971-974.
30. Kanadia,R.N., Urbinati,C.R., Crusselle,V.J., Luo,D., Lee,Y.J., Harrison,J.K., Oh,S.P., and
Swanson,M.S. (2003) Developmental expression of mouse muscleblind genes Mbnl1, Mbnl2 and
Mbnl3. Gene Expr.Patterns., 3, 459-462.
31. Jansen,G., Groenen,P.J., Bachner,D., Jap,P.H., Coerwinkel,M., Oerlemans,F., van den,B.W.,
Gohlsch,B., Pette,D., Plomp,J.J. et al. (1996) Abnormal myotonic dystrophy protein kinase levels
produce only mild myopathy in mice. Nat.Genet., 13, 316-324.
32. Reddy,S., Smith,D.B., Rich,M.M., Leferovich,J.M., Reilly,P., Davis,B.M., Tran,K., Rayburn,H.,
Bronson,R., Cros,D. et al. (1996) Mice lacking the myotonic dystrophy protein kinase develop a
late onset progressive myopathy. Nat.Genet., 13, 325-335.
33. Mankodi,A., Logigian,E., Callahan,L., McClain,C., White,R., Henderson,D., Krym,M., and
Thornton,C.A. (2000) Myotonic dystrophy in transgenic mice expressing an expanded CUG
repeat. Science, 289, 1769-1773.
34. Seznec,H., Agbulut,O., Sergeant,N., Savouret,C., Ghestem,A., Tabti,N., Willer,J.C., Ourth,L.,
Duros,C., Brisson,E. et al. (2001) Mice transgenic for the human myotonic dystrophy region with
expanded CTG repeats display muscular and brain abnormalities. Hum.Mol.Genet., 10, 2717-
2726.
35. Schulz,P.E., McIntosh,A.D., Kasten,M.R., Wieringa,B., and Epstein,H.F. (2003) A role for
myotonic dystrophy protein kinase in synaptic plasticity. J.Neurophysiol., 89, 1177-1186.
36. Davies,S.W., Turmaine,M., Cozens,B.A., DiFiglia,M., Sharp,A.H., Ross,C.A., Scherzinger,E.,
Wanker,E.E., Mangiarini,L., and Bates,G.P. (1997) Formation of neuronal intranuclear inclusions
underlies the neurological dysfunction in mice transgenic for the HD mutation. Cell, 90, 537-548.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
37. DiFiglia,M., Sapp,E., Chase,K.O., Davies,S.W., Bates,G.P., Vonsattel,J.P., and Aronin,N. (1997)
Aggregation of huntingtin in neuronal intranuclear inclusions and dystrophic neurites in brain.
Science, 277, 1990-1993.
38. Haltia,M., Somer,H., Palo,J., and Johnson,W.G. (1984) Neuronal intranuclear inclusion disease in
identical twins. Ann.Neurol., 15, 316-321.
39. Greco,C.M., Hagerman,R.J., Tassone,F., Chudley,A.E., Del Bigio,M.R., Jacquemont,S.,
Leehey,M., and Hagerman,P.J. (2002) Neuronal intranuclear inclusions in a new cerebellar
tremor/ataxia syndrome among fragile X carriers. Brain, 125, 1760-1771.
40. Klement,I.A., Skinner,P.J., Kaytor,M.D., Yi,H., Hersch,S.M., Clark,H.B., Zoghbi,H.Y., and
Orr,H.T. (1998) Ataxin-1 nuclear localization and aggregation: role in polyglutamine-induced
disease in SCA1 transgenic mice. Cell, 95, 41-53.
41. Sisodia,S.S. (1998) Nuclear inclusions in glutamine repeat disorders: are they pernicious,
coincidental, or beneficial? Cell, 95, 1-4.
42. Liquori,C.L., Ricker,K., Moseley,M.L., Jacobsen,J.F., Kress,W., Naylor,S.L., Day,J.W., and
Ranum,L.P. (2001) Myotonic dystrophy type 2 caused by a CCTG expansion in intron 1 of ZNF9.
Science, 293, 864-867.
43. van der Ven,P.F., Jansen,G., van Kuppevelt,T.H., Perryman,M.B., Lupa,M., Dunne,P.W., ter
Laak,H.J., Jap,P.H., Veerkamp,J.H., Epstein,H.F. et al. (1993) Myotonic dystrophy kinase is a
component of neuromuscular junctions. Hum.Mol.Genet., 2, 1889-1894.
44. Whiting,E.J., Waring,J.D., Tamai,K., Somerville,M.J., Hincke,M., Staines,W.A., Ikeda,J.E., and
Korneluk,R.G. (1995) Characterization of myotonic dystrophy kinase (DMK) protein in human
and rodent muscle and central nervous tissue. Hum.Mol.Genet., 4, 1063-1072.
45. Pham,Y.C., Man,N., Lam,L.T., and Morris,G.E. (1998) Localization of myotonic dystrophy
protein kinase in human and rabbit tissues using a new panel of monoclonal antibodies.
Hum.Mol.Genet., 7, 1957-1965.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
46. Lam,L.T., Pham,Y.C., Nguyen,T.M., and Morris,G.E. (2000) Characterization of a monoclonal
antibody panel shows that the myotonic dystrophy protein kinase, DMPK, is expressed almost
exclusively in muscle and heart. Hum.Mol.Genet., 9, 2167-2173.
47. Sarkar,P.S., Han,J., and Reddy,S. (2004) In situ hybridization analysis of Dmpk mRNA in adult
mouse tissues. Neuromuscul.Disord., 14, 497-506.
48. Pal,R., Agbas,A., Bao,X., Hui,D., Leary,C., Hunt,J., Naniwadekar,A., Michaelis,M.L.,
Kumar,K.N., and Michaelis,E.K. (2003) Selective dendrite-targeting of mRNAs of NR1 splice
variants without exon 5: identification of a cis-acting sequence and isolation of sequence-binding
proteins. Brain Res., 994, 1-18.
49. Durand,G.M., Bennett,M.V., and Zukin,R.S. (1993) Splice variants of the N-methyl-D-aspartate
receptor NR1 identify domains involved in regulation by polyamines and protein kinase C.
Proc.Natl.Acad.Sci.U.S.A, 90, 6731-6735.
50. Traynelis,S.F., Hartley,M., and Heinemann,S.F. (1995) Control of proton sensitivity of the NMDA
receptor by RNA splicing and polyamines. Science, 268, 873-876.
51. Tsien,J.Z., Huerta,P.T., and Tonegawa,S. (1996) The essential role of hippocampal CA1 NMDA
receptor-dependent synaptic plasticity in spatial memory. Cell, 87, 1327-1338.
52. Tonegawa,S., Tsien,J.Z., McHugh,T.J., Huerta,P., Blum,K.I., and Wilson,M.A. (1996)
Hippocampal CA1-region-restricted knockout of NMDAR1 gene disrupts synaptic plasticity,
place fields, and spatial learning. Cold Spring Harb.Symp.Quant.Biol., 61, 225-238.
53. Portwood,M.M., Wicks,J.J., Lieberman,J.S., and Duveneck,M.J. (1986) Intellectual and cognitive
function in adults with myotonic muscular dystrophy. Arch.Phys.Med.Rehabil., 67, 299-303.
54. Rubinsztein,J.S., Rubinsztein,D.C., McKenna,P.J., Goodburn,S., and Holland,A.J. (1997) Mild
myotonic dystrophy is associated with memory impairment in the context of normal general
intelligence. J.Med.Genet., 34, 229-233.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
55. Goedert,M., Spillantini,M.G., Jakes,R., Rutherford,D., and Crowther,R.A. (1989) Multiple
isoforms of human microtubule-associated protein tau: sequences and localization in
neurofibrillary tangles of Alzheimer's disease. Neuron, 3, 519-526.
56. Hong,M., Zhukareva,V., Vogelsberg-Ragaglia,V., Wszolek,Z., Reed,L., Miller,B.I.,
Geschwind,D.H., Bird,T.D., McKeel,D., Goate,A. et al. (1998) Mutation-specific functional
impairments in distinct tau isoforms of hereditary FTDP-17. Science, 282, 1914-1917.
57. Goedert,M., Jakes,R. (1990) Expression of separate isoforms of human tau protein: correlation
with the tau pattern in brain and effects on tubulin polymerization. EMBO J., 9, 4225-4230.
58. Hutton,M., Lendon,C.L., Rizzu,P., Baker,M., Froelich,S., Houlden,H., Pickering-Brown,S.,
Chakraverty,S., Isaacs,A., Grover,A. et al. (1998) Association of missense and 5'-splice-site
mutations in tau with the inherited dementia FTDP-17. Nature, 393, 702-705.
59. Spillantini,M.G., Murrell,J.R., Goedert,M., Farlow,M.R., Klug,A., and Ghetti,B. (1998) Mutation
in the tau gene in familial multiple system tauopathy with presenile dementia.
Proc.Natl.Acad.Sci.U.S.A, 95, 7737-7741.
60. Poorkaj,P., Bird,T.D., Wijsman,E., Nemens,E., Garruto,R.M., Anderson,L., Andreadis,A.,
Wiederholt,W.C., Raskind,M., and Schellenberg,G.D. (1998) Tau is a candidate gene for
chromosome 17 frontotemporal dementia. Ann.Neurol., 43, 815-825.
61. D'Souza,I., Poorkaj,P., Hong,M., Nochlin,D., Lee,V.M., Bird,T.D., and Schellenberg,G.D. (1999)
Missense and silent tau gene mutations cause frontotemporal dementia with parkinsonism-
chromosome 17 type, by affecting multiple alternative RNA splicing regulatory elements.
Proc.Natl.Acad.Sci.U.S.A, 96, 5598-5603.
62. Lee,V.M., Goedert,M., and Trojanowski,J.Q. (2001) Neurodegenerative tauopathies.
Annu.Rev.Neurosci., 24, 1121-1159.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
63. Stanford,P.M., Shepherd,C.E., Halliday,G.M., Brooks,W.S., Schofield,P.W., Brodaty,H.,
Martins,R.N., Kwok,J.B., and Schofield,P.R. (2003) Mutations in the tau gene that cause an
increase in three repeat tau and frontotemporal dementia. Brain, 126, 814-826.
64. Foster,N.L., Wilhelmsen,K., Sima,A.A., Jones,M.Z., D'Amato,C.J., and Gilman,S. (1997)
Frontotemporal dementia and parkinsonism linked to chromosome 17: a consensus conference.
Ann.Neurol., 41, 706-715.
65. Kiuchi,A., Otsuka,N., Namba,Y., Nakano,I., and Tomonaga,M. (1991) Presenile appearance of
abundant Alzheimer's neurofibrillary tangles without senile plaques in the brain in myotonic
dystrophy. Acta Neuropathol., 82, 1-5.
66. Yoshimura,N., Otake,M., Igarashi,K., Matsunaga,M., Takebe,K., and Kudo,H. (1990) Topography
of Alzheimer's neurofibrillary change distribution in myotonic dystrophy. Clin.Neuropathol., 9,
234-239.
67. Vermersch,P., Sergeant,N., Ruchoux,M.M., Hofmann-Radvanyi,H., Wattez,A., Petit,H.,
Dwailly,P., and Delacourte,A. (1996) Specific tau variants in the brains of patients with myotonic
dystrophy. Neurology, 47, 711-717.
68. Sergeant,N., Sablonniere,B., Schraen-Maschke,S., Ghestem,A., Maurage,C.A., Wattez,A.,
Vermersch,P., and Delacourte,A. (2001) Dysregulation of human brain microtubule-associated tau
mRNA maturation in myotonic dystrophy type 1. Hum.Mol.Genet., 10, 2143-2155.
69. Ishihara,T., Zhang,B., Higuchi,M., Yoshiyama,Y., Trojanowski,J.Q., and Lee,V.M. (2001) Age-
dependent induction of congophilic neurofibrillary tau inclusions in tau transgenic mice.
Am.J.Pathol., 158, 555-562.
70. Probst,A., Gotz,J., Wiederhold,K.H., Tolnay,M., Mistl,C., Jaton,A.L., Hong,M., Ishihara,T.,
Lee,V.M., Trojanowski,J.Q. et al. (2000) Axonopathy and amyotrophy in mice transgenic for
human four-repeat tau protein. Acta Neuropathol., 99, 469-481.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
71. Artero,R., Prokop,A., Paricio,N., Begemann,G., Pueyo,I., Mlodzik,M., Perez-Alonso,M., and
Baylies,M.K. (1998) The muscleblind gene participates in the organization of Z-bands and
epidermal attachments of Drosophila muscles and is regulated by Dmef2. Dev.Biol., 195, 131-143.
72. Ho,T.H., Charlet,B., Poulos,M.G., Singh,G., Swanson,M.S., and Cooper,T.A. (2004) Muscleblind
proteins regulate alternative splicing. EMBO J., 23, 3103-3112.
73. Ladd,A.N., Charlet,N., and Cooper,T.A. (2001) The CELF family of RNA binding proteins is
implicated in cell-specific and developmentally regulated alternative splicing. Mol.Cell Biol., 21,
1285-1296.
74. Lu,X., Timchenko,N.A., and Timchenko,L.T. (1999) Cardiac elav-type RNA-binding protein
(ETR-3) binds to RNA CUG repeats expanded in myotonic dystrophy. Hum.Mol.Genet., 8, 53-60.
75. Michalowski,S., Miller,J.W., Urbinati,C.R., Paliouras,M., Swanson,M.S., and Griffith,J. (1999)
Visualization of double-stranded RNAs from the myotonic dystrophy protein kinase gene and
interactions with CUG-binding protein. Nucleic Acids Res., 27, 3534-3542.
76. Mankodi,A., Teng-Umnuay,P., Krym,M., Henderson,D., Swanson,M., and Thornton,C.A. (2003)
Ribonuclear inclusions in skeletal muscle in myotonic dystrophy types 1 and 2. Ann.Neurol., 54,
760-768.
77. Zhang,W., Liu,H., Han,K., and Grabowski,P.J. (2002) Region-specific alternative splicing in the
nervous system: implications for regulation by the RNA-binding protein NAPOR. RNA., 8, 671-
685.
78. Wang,J., Gao,Q.S., Wang,Y., Lafyatis,R., Stamm,S., and Andreadis,A. (2004) Tau exon 10,
whose missplicing causes frontotemporal dementia, is regulated by an intricate interplay of cis
elements and trans factors. J.Neurochem., 88, 1078-1090.
79. Good,P.J., Chen,Q., Warner,S.J., and Herring,D.C. (2000) A family of human RNA-binding
proteins related to the Drosophila Bruno translational regulator. J.Biol.Chem., 275, 28583-28592.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
80. Dreyfuss,G., Kim,V.N., and Kataoka,N. (2002) Messenger-RNA-binding proteins and the
messages they carry. Nat.Rev.Mol.Cell Biol., 3, 195-205.
81. Min,H., Chan,R.C., and Black,D.L. (1995) The generally expressed hnRNP F is involved in a
neural-specific pre-mRNA splicing event. Genes Dev., 9, 2659-2671.
82. Chou,M.Y., Rooke,N., Turck,C.W., and Black,D.L. (1999) hnRNP H is a component of a splicing
enhancer complex that activates a c-src alternative exon in neuronal cells. Mol.Cell Biol., 19, 69-
77.
83. Cummings,C.J., Mancini,M.A., Antalffy,B., DeFranco,D.B., Orr,H.T., and Zoghbi,H.Y. (1998)
Chaperone suppression of aggregation and altered subcellular proteasome localization imply
protein misfolding in SCA1. Nat.Genet., 19, 148-154.
84. Mitchell,P., Petfalski,E., Shevchenko,A., Mann,M., and Tollervey,D. (1997) The exosome: a
conserved eukaryotic RNA processing complex containing multiple 3'-->5' exoribonucleases. Cell,
91, 457-466.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
FIGURE LEGENDS
Figure 1. FISH (left panels) and immunofluorescence (IF, middle panels) on frozen sections of DM1
brain shows nuclear foci of mutant DMPK mRNA. FISH, IF, and nuclear stain (DAPI, blue) images are
merged in panels on the right. A. FISH (without IF) using Texas Red-labeled CAG repeat probe shows
an RNA inclusion in frontal cortical neuron. Autofluorescence from lipofuscin occurs at broad spectrum
of wavelengths. It appears in every color channel and as yellow-brown perinuclear material in the merged
image. RNA inclusions in cerebral cortex are confined to neurons identified by IF for NeuN (B) or
MAP2 (C). Small foci are present in cerebellar Purkinje cells (D) or oligodendrocytes of the centrum
semiovale (E) identified by IF for calbindin or CNPase, respectively. (F) RNA foci do not colocalize
with PML bodies in cortical neurons. Bar, 5 µm, applies to all panels.
Figure 2. FISH and IF on sections of temporal or frontal cortical neurons show colocalization of mutant
DMPK mRNA [(CUG)n] with 20Sα subunit of proteasome (A), MBNL2 (E), and hnRNP F (F). There is
a marked redistribution of MBNL1 into RNA foci in DM1 cortical neurons (G), compared to the
distribution in the nucleus (excluding nucleolus) and cytoplasm of normal neurons (H). Mutant DMPK
mRNA does not colocalize with the PM/Scl100 (nuclear) component of the exosome (B), CUGBP1 (C),
or NF90 (D). RARγ does not colocalize with RNA foci in DM1 cortical neurons (I). The distribution of
RARγ in the DM1 (I) and non-neurologic-disease (J) neuronal nucleus is similar. Bar, 5 µm, applies to
all panels.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Figure 3. MBNL1 is decreased in nucleoplasm of DM1 cortical neurons. Immunofluorescence (area ×
intensity) for MBNL1 in the nucleus, excluding nucleolus and RNA foci, was determined for 20 neurons
in sections of temporal cortex from 3 individuals with DM1 and 3 controls without neurologic disease
(C).
Figure 4. Alternative splicing of NMDA NR1 receptor (NMDAR1), amyloid beta precursor protein
(APP), and microtubule-associated protein tau (MAPT) is abnormally regulated in DM1. A. Splice
products obtained by RT-PCR amplification of RNA isolated from non-disease control (n=5) or DM1
(n=7) temporal cortex. Exon utilization for each splice product is shown in diagram. B. Quantification of
RT-PCR splicing assay (triplicates). ex, exon.
Figure S1. RNA foci in dentate gyrus and subcortical neurons in DM1. FISH (CAG repeat probe, red)
merged with IF (anti-NeuN antibody, green) and nuclear stain (DAPI, blue). SN, substantia nigra. Bar, 5
µm, applies to all panels.
Figure S2. Foci of mutant RNA in neuronal and muscle nuclei. A. When processed on the same slide
and imaged under the same exposure settings, RNA foci in frontal cortex were larger and more intensely
fluorescent than in biceps muscle from the same patient. B. In paired samples of frontal cortex and
skeletal muscle from the same patient, fluorescence area × intensity of RNA foci was 3.1-fold greater in
nuclei of cortical neurons than in skeletal muscle, n=3 patients, 20 nuclei per sample (p<10-10).
Figure S3. FISH combined with IF on sections of DM1 temporal or frontal cortex. FISH (CAG repeat
probe, red, left panels) and IF (middle panels, antibody to indicated protein, green) are merged with
nuclear stain (DAPI, blue) in right panels. CUG expansion RNA colocalizes with proteasome 11Sγ
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
subunit (A) and hnRNP H (C) but not with double-stranded RNA binding protein ADAR1 (B), hnRNP M
(D), or Sp1 (E). Bar, 5 µm, applies to all panels.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Supplemental table 1
Target antigen Final Dilution Source
muscleblind
(MBNL1)
mAb (3B10): 1:1500
pAb (EXP 42) : 1:1500
M. Swanson, FL
muscleblind
(MBNL2)
mAb (2D9): 1:10,000
M. Swanson, FL
CUGBP1 mAb (3B1): 1:500 M. Swanson, FL
CELF4 pAb (#440): 1:500
T. Cooper, TX
ETR3 pAb ( #163): 1:500
T. Cooper, TX
PKR pAb (pT451): 1:500
pAb (M515):1:500
pAb (D20): 1:500
mAb (B10): 1:500
Biosource International,CA
Santa Cruz, CA
Santa Cruz, CA
Santa Cruz, CA
RNA helicase A pAb: 1:500
C. Lee, NJ
ADAR1 mAb: 1:500 D. Cho, PA
HRBP pAb (#1683):1:500
C. Thornton
NF90 (DRBP76) pAb (p90 AB4): 1:500
G. Sen, OH
Staufen pAb 1:500 (AB 5819)
Chemicon, Temecula, CA
Proteosome
19S S10a
11S α
11S γ
20S β3 (HC10)
20S α
pAb: 1:500
pAb: 1:1000
pAb: 1:1000
mAb: 1:500
pAb: 1:1000
Affiniti, Exeter, UK
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Ubiquitin pAb: 1:1000 DAKO, Glostrup, Denmark
p80 coilin pAb (R288):1:500 KL Chan, CA
C23 nucleolin mAb: 1:500
Santa Cruz, CA
PML pAb: 1:500
Santa Cruz, CA
PTB pAb: 1:500
E. Wagner, NC
PM-Scl 75 pAb: 1:500
J. Wilusz, Newark
hnRNP H
C-terminal
N-terminal
pAb: 1:1000
pAb: 1: 500
J. Wilusz, Newark
hnRNP F pAb: 1:1000
J. Wilusz, Newark
hnRNP H
hnRNP F
hnRNPI /PTB
KSRP
4F4 (hnRNP C)
1D8 (hnRNP M)
pAb: 1:1000
pAb: 1:1000
pAb: 1:1000
pAb: 1:1000
mAb: 1:500
mAb: 1:500
Doug Black, CA
M. Swanson, FL
CNPase
mAb: 1:1000 Sigma, St. Louis, MO
MAP2 mAb: 1:500
Sigma, St. Louis, MO
NeuN mAb: 1:500
Chemicon, Temecula, CA
Sp1 (sc-59) pAb: 1:500
Santa Cruz Biotechnology, Santa Cruz,
CA
RARγ(sc-550) pAb: 1:500
Santa Cruz Biotechnology, Santa Cruz,
CA
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Staufen (AB5819) pAb: 1:500
Chemicon, Temecula, CA92590
SUMO-1 mAb
Zymed Laboratories Inc., South San
Francisco. CA94080
Supplemental Table 1. Sources for primary antibodies. mAb, monoclonal antibody; pAb, polyclonal
antibody.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Supplemental table 2
Gene Name Unigene
Alternatively
spliced exon Acc. No. Nucleotides
Amyloid beta (A4) precursor protein APP ex2 NM_000484 205-372
ex7 NM_000484 1013-1180
ex15 NM_000484 1181-1237
Actin-related protein 3-beta ARP3BETA ex2 BC008682.1 134-189
Beta-site APP-cleaving enzyme 2 BACE2 ex9 NM_012105 1448-1597
ex10 NM_012105 1598-1766
Neuronal apoptosis inhibitory protein BIRC1 ex10-11 NM_004536 1314-1453
Calcium channel, voltage-dependent, P/Q
type, alpha 1A subunit
CACNA1A ex38 AF004883 5783-5879
Calcium/calmodulin protein kinase II
dependent delta
CAMK2D alt splic donor in
ex21(44bp)
NM_172127 1709-1981
Clathrin light chain B CLTB exon casette NM_007097 641-694
Homo sapiens discs, large homolog 1 DLG1 ex8 NM_004087 771-824
Dopamine receptor type 2 DRD2 ex6 NM_000795 889-975
Erythrocyte membrane protein band 4.1-like 1 EPB41L1 ex5 NM_012156 514-618
ex21 NM_012156 2356-2439
GABA receptor alpha 2 GABRA4 ex3-8 NM_000809 345-1274
Gephyrin GPHN ex9 NM_020806 742-840
ex12 NM_020806 976-1018
LIM domain binding 3 LDB3 ex10 AB 014513 918-1106
NMDA receptor NR1 GRIN1 ex5 AF015730 644-706
ex20 NM_007327 3683-3793
ex21 NM_007327 3794-3910
Neurotrophic tyrosine kinase, receptor, type 2 NTRK2 exon casette AF410901 1878-1926
C-Jun N-terminal kinase 2 JNK2 E6b, E6a NM_002752 666-737
Netrin G1 Ntng2 E5 NM_032536 1115-1138
Neogenin NEO1 ex26 NM_002499 3879-4037
Neurofibromin 1 NF1 ex9a NT010799 108004-108033
Neuronatin NNAT ex2 NM_005386 200-280
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Neurorexin 1 NRXN1 ex3a NM_004801 965-1024
ex4
ex5
ex7a NM_004801 1327-1250
ex12 NM_004801 2540-2566
Neurorexin 2 NRXN2 ex12 NM_015080 2829-2855
ex20 NM_015080 4197-4286
Neurorexin 3 NRXN3 ex12 NM_004796 1621-1647
NUMB NUMB ex8 AF015040 480-512
ex15 AF015040 1367-1510
Presynaptic cytomatrix protein PICO ex10 AB011131 3056-3082
Peanut-like 2 PNUTL2 ex2 NM_004574 189-579
Protein phosphatase 2, regulatory subunit B
(PR 52), beta isoform
PPP2R2B ex6 MN_181674 364-557
REST/NRSF/SBR REST ex5 AF228045 410-459
Microtubule-associated protein tau MAPT ex2 NM_005910 370-456
ex10 NM_005910 1059-1151
Sarcolemma associated protein SLMAP ex4 NM_007159 273-395
v-src sarcoma (Schmidt-Ruppin A-2) viral
oncogene homolog (avian)
SRC exon casette NM_005417 additional exon
cassette (18bp or
50bp)
Supplemental table 2. List of exons screened for abnormal regulation of alternative splicing in DM1
compared to controls without neurologic disease. “Nucleotides” indicates which portion of the specified
cDNA (GenBank accession number) was amplified by RT-PCR.
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Figure 1
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Figure 2
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Figure 3
C C C DM1 DM1 DM1
MB
NL1
in n
ucle
opla
sm(fl
uore
scen
ce in
tens
ity x
are
a)
0
20
40
60
80
100
120
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Figure 4
4 5 64 6
6 7 96 9
1 2 4
1 4
9 10 119 11
APP
NMDAR1
MAPT
MAPT
4 5 64 64 5 64 5 64 64 6
6 7 96 7 96 96 9
1 2 41 2 4
1 41 4
9 10 119 10 119 119 11
APP
NMDAR1
MAPT
MAPT
A
C DM1 C DM1 C DM1 C DM1 APP NMDAR1 MAPT MAPT ex7 ex5 ex2 ex10
% in
clus
ion
0
10
20
30
40
50
60B
C DM1 C DM1 C DM1 C DM1 APP NMDAR1 MAPT MAPT ex7 ex5 ex2 ex10
% in
clus
ion
0
10
20
30
40
50
60Bcontrol DM1
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Supplemental figure 1
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Supplemental figure 2
0
10000
20000
Cortical Neurons Sk. Muscle
Cortical neuron Biceps muscle
Fluo
resc
ence
(are
a ×
inte
nsity
)
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from
Supplemental figure 3
by guest on May 30, 2013
http://hmg.oxfordjournals.org/
Dow
nloaded from